DESCRIPTION: (Applicant's Description) Preoperative chemotherapy is now used commonly in patients with advanced disease. In other patients, post-operative chemotherapy has been viewed as essential to prevent metastatic spread. In many cases, however, the success rate for chemotherapy is relatively poor. Recent laboratory studies have shown that the tumor suppressor protein p53 causes cellular death in response to DNA damaging agents. The most commonly used chemotherapeutic drugs in breast cancer exert their effects through DNA damage and thus it is expected that the p53-dependent pathway is central to the responsiveness of breast tumors to these agents. We hypothesize that certain components of the p53-dependent pathway in human breast tumors determine individual responsiveness to different chemotherapeutic regimens. We also hypothesize that mutant p53 does not appropriately mediate the response to chemotherapy as does wild-type p53. Our data have shown that human breast tumor cell lines with missense mutation of the p53 gene or loss of p53 expression are not growth arrested by doxorubicin, methotrexate, or 5-fluorouracil. This research is driven by the notion that it is inactivation of a p53-dependent pathway rather than genetic mutation of p53 per se which results in reduced responsiveness to chemotherapeutic agents which mediate their effects via DNA damage. The goals of this proposal are to (1) Determine whether the p53 is mutant in 60 patients who received a chemotherapeutic regimen involving DNA damaging agents through sequencing of genomic DNA. (2) Characterize the expression level of proteins which may cause p53 dysfunction in these tumor samples, namely mdm2 and cyclin D1, by immunohistochemistry. (3) Determine whether there is an association between p53 status (mutant versus wild-type), the levels of these regulators, and the responsiveness of the patients to chemotherapy with a standard regimen of cyclophosphamide, doxorubicin, and 5-fluorouracil (CAF). The long term goal of such studies is to determine if the presence of mutant p53 and/or overexpression of mdm2 or cyclin D1 determines the successful use of a particular chemotherapeutic regimen. Thus, patients will be able to tailor chemotherapy to their own biological response. Ideally, such studies will lead to designing appropriate chemotherapeutic regimens for particular genetically-defined tumors allowing for more efficacious eradication of potential metastatic disease. The long term goal, then, is to demonstrate that tumor genotype is a useful predictor of chemotherapeutic responsiveness.